Where to add the thickness of a stiffner in a partial-depth end-plate calculation. 4

[As-Lag]

Dear Engineers

I am designing a partial-depth end-plate for a beam that is double-notched and is carrying High Shear across the notched section and want to add a plate to the web of the supported beam to increase its cross-sectional area.

I am using the Eurocodes and the check I am performing on my beam may be refered to as 'Check 5' of the SCI-P358 publication - Page 28 and 29. I have yet to find a publication detailing the requirement of stiffening a beam (offers of links to publications are always welcome) so I am unsure on how to include the stiffner in my calculations.

My questions are:

Do I simply add the cross-section area of the stiffner resisting shear to the 'Shear Area' of the calculation?; and for the weld: how would I proportion the shear-action in the welds of the stiffner? I am guessing the stresses would be shared in proportion to the individual plates shear-areas.

Regards - Dan

 

[271828]

Are you adding a stiffener -- plate with large dim perpendicular to the web? Or a doubler plate with large dim parallel to the web?

The AISC Companion to the Steel Manual has an example for the latter at least in the Shear Connections part. It's a free download at the website.

 

[centondollar]

I have yet to find a publication detailing the requirement of stiffening a beam (offers of links to publications are always welcome) so I am unsure on how to include the stiffner in my calculations.

You stiffen the web to prevent shear buckling, and if the web is used to resist bending moment, it may on rare occasions require longitudinal stiffeners to prevent compression buckling. End-posts (rigid or non-rigid, one-sided or two-sided, welded to both upper and lower flanges) can be used at supports and at notches to provide a rigid diaphragm, which in turn anchors the shear buckling tension field.

Adding stiffener areas to shear area is not something I've heard of, and if you think about the mechanics, it doesn't make much sense. The shear force varies continuously along the beam, while the added shear area of the stiffener is active only for a very short distance (practically only for the stiffener thickness), and is therefore not effective in picking up shear. The same argument applies to torsional stiffness of I-beams with stiffeners: it is not noticeably increased unless stiffeners are continuous and turn the beam into a closed box-type section.

 

[As-Lag]

Thanks for the replies.

I attach a snip showing my stiffner.

Regards Dan

 

[271828]

In AISC Land, that's called a doubler plate. The Companion to the AISC Steel Construction Manual Volume 1: Design Examples Example II.A-6 covers this.

Free download at:

https://www.aisc.org/publications/steel-construction-manual-resources/#37583

 

[As-Lag]

271828

Thanks for the link and I have downloaded the document. Apart from the imperial measurements, I now have an idea on how to proceed. I need to get my head around the horizontal stiffner too as it may be more suitable than a doubler-plate.

Regards

 

[271828]

Stiffeners would increase the bending strength. It wouldn't do anything for shear.

 

[SWComposites]

And that stiffener is going to put a stress concentration right in the gap between the stiffener and doubler plate, where you don’t want it. And agree, the stiffener does nothing for the shear load.

The doubler detail looks ugly; probably should be connected to the beam flanges.

 

[human909]

I think most people here have responded well and pretty much covered most things.

If you really won't to go deeper on this topic this is also a good resource. One of the authors is our good old mate Joseph Yura who be the GOD of LTB research in steel beams.

https://files.engineering.com/getfi...&file=DESIGN_AND_BEHAVIOUR_OF_COPED_BEAMS.pdf

 

[As-Lag]

Thanks chaps.

My notched beam is failing in bending, not shear. The shear force is applied though the cut area and has an eccentricity causing bending (Check 5 of the SCI document - see above).

I cannot see why a doubler-plate is wrong. The method is shown in many books but with little science to validate the method.

Without other evidence, I am going to assume the plates reinforce the web against tearing and apply a rotation to the plate and welds due to the moment and check the welds. The plates would effectivelly increase the thickness of the web.

OR, is there a better connection that would be able to accept the loadings applied? Fin plate maybe.

Regards

 

[271828]

The AISC example covers flexure also. The stiffener or doubler needs to extend into the uncoped portion of the beam some distance.

Because of that extension, the high toughness of mild steel, proper detailing with radii instead of sharp corners in tension, we're not concerned with stress concentrations / fracture due the addition of the stiffeners or doublers. Such reinforcement happens with some regularity, so this isn't a "weird" problem. Just to hopefully make you feel better about it. LOL

 

[KootK]

I cannot see why a doubler-plate is wrong.

It's not wrong. It's probably just not well described as a stiffener or a doubler as those terms have specific meanings in the literature which are a bit different from what you're trying to do. I'd be inclined to call it a "flexural extension plate". And you can make a go of it so long as the plates work for bending, shear, and stability and the fastener group (bolts or welds) can successfully moment connect the plates to the uncoped portions of the beam web.

OR, is there a better connection that would be able to accept the loadings applied? Fin plate maybe.

1) My first choice would be the extended shear tab configuration. You can find out a bit about that here: Link. It eliminates the need for beam coping altogether and is a method included in the latest AISC manual.

2) If you decide to stick with your original scheme, I'd replace your double plates with a single channel. That, simply because channels handle bending and local torsional stability better than plates do. I couldn't find a good photo of it but there are a bunch of beam in Seattle's public library that are end supported this way.